IMAGE COMPRESSION USING MULTILAYER NEURAL NETWORKS

A new neural-network data compression method is presented. The work ex tends the use of two-layer neural networks to multilayer networks, Res ults show the performance superiority of multilayer neural networks co mpared with that of the two-layer one, especially at high compression ratios. To overcome the long training time required for multilayered n etworks, a recently developed training algorithm has been used. A modf ied feedback error is proposed to reduce further the training time and to enhance the image quality. Also, a redistribution of the grey leve ls in the training phase is proposed to make the minimisation of the m ean-square error more related to the human-vision system.